Detecting portions of interest in images

Abstract

A computer-implemented method of automatically locating a portion of interest in image or matrix data derived from an item under consideration includes: identifying parts of the image or matrix data that satisfy a preset threshold as objects which are possibly parts of the portion of the interest; applying at least one preselected filter to the data corresponding to the objects to find a set of objects consisting of the objects most likely to be part of the portion of interest; sorting the objects of the set into clusters according to a predefined criterion; and using a known characteristic of the portion of interest to identify which one of the clusters corresponds to the portion of interest.

What is claimed is: 
     
       1. A computer-implemented method of automatically locating a portion of interest in one of an image and matrix data, derived from an item under consideration, comprising:
 identifying parts of the one of the image and the matrix data that satisfy a preset threshold as identified objects possibly included in the portion of interest, the portion of interest corresponding to one of a layer and a boundary, and one of representing and forming part of the item under consideration; 
 applying at least one preselected filter to the one of the image and the matrix data corresponding to the identified objects to find a set of interesting objects most likely to be part of the portion of interest, the preselected filter filtering out the identified objects having a major axis in a direction perpendicular to an expected orientation of the one of the layer and the boundary; 
 sorting the interesting objects in the set into clusters according to a predefined criterion; and 
 identifying, in the one of the image and the matrix data based on a known characteristic of the portion of interest, one of the clusters as corresponding to the portion of interest. 
 
     
     
       2. A method as claimed in  claim 1 , wherein, for image data, the preset threshold comprises a pixel value for at least one color, and pixel data of the image which exceed the pixel value are included in the identified objects. 
     
     
       3. A method as claimed in  claim 1 , wherein the predefined criterion for sorting the interesting objects in the set is position in the direction perpendicular to the expected orientation of the one of the layer and the boundary. 
     
     
       4. A method as claimed in  claim 1 , further comprising, generating an image from the matrix data excluding cluster data of at least one cluster not identified as corresponding to the portion of interest. 
     
     
       5. A method as claimed in  claim 1 , further comprising, removing from the image, image data of at least one cluster not identified as corresponding to the portion of interest. 
     
     
       6. A method as claimed in  claim 1 , wherein the one of the image and matrix data is derived from ultrasound scan data of the item under consideration. 
     
     
       7. A method as claimed in  claim 1 , wherein the item under consideration is at least one of: a manufactured item; and an item used in manufacture. 
     
     
       8. A defect detection method for use in detecting a defect in an item under consideration, comprising:
 analyzing one of an image of the item and an image derived from the item, to identify the defect; and 
 obtaining image data for the analyzing in which a portion of interest has been automatically located in accordance with the method of  claim 1 . 
 
     
     
       9. A method as claimed in  claim 8 , comprising identifying, as a site of a potential defect, a location of a gap in the data forming the cluster identified as corresponding to the portion of interest. 
     
     
       10. A method as claimed in  claim 8 , wherein the item under consideration is at least one of: a manufactured item; and an item used in manufacture. 
     
     
       11. A method as claimed in  claim 8 , further comprising determining at least one of: structural integrity of material forming at least part of the item; and flaws in material forming at least part of the item. 
     
     
       12. A non-transitory computer-readable storage medium carrying instructions which, when run on a computer, causes that computer to carry out a method as claimed in  claim 1 . 
     
     
       13. A method as claimed in  claim 1 , further comprising, after identifying the portion of interest, analyzing the portion of interest to identify a defect in the item under consideration. 
     
     
       14. A method as claimed in  claim 8 , further comprising presenting the portion of interest having the defect to a user. 
     
     
       15. A method as claimed in  claim 1 , wherein the one of the image and matrix data is stored, and the method further comprises, after identifying the portion of interest, reducing a stored amount of the one of the image and matrix data by eliminating part of the data in accordance with a location of the identified portion of interest. 
     
     
       16. A data processing apparatus to locate a portion of interest in one of image and matrix data, derived from an item under consideration, the apparatus comprising:
 a processor; and 
 a memory storing instructions that when executed by the processor
 identify parts of the one of the image and the matrix data that satisfy a preset threshold as identified objects possibly included in the portion of interest, the portion of interest corresponding to one of a layer and a boundary, and one of representing and forming part of the item under consideration, 
 apply at least one preselected filter to the one of the image and the matrix data corresponding to the identified objects to find a set of interesting objects most likely to be part of the portion of interest, the preselected filter filtering out the identified objects having a major axis in a direction perpendicular to an expected orientation of the one of the layer and the boundary, 
 sort the interesting objects in the set into clusters according to a predefined criterion, and 
 identify, in the one of the image and the matrix data based on a known characteristic of the portion of interest, one of the clusters as corresponding to the portion of interest. 
 
 
     
     
       17. An apparatus as claimed in  claim 16 , wherein the instructions further include to identify, as a site of a potential defect, a location of a gap in cluster data forming the cluster identified as corresponding to the portion of interest. 
     
     
       18. An apparatus as claimed in  claim 16 , further comprising a display to display the portion of interest. 
     
     
       19. A data processing apparatus as claimed in  claim 16 , wherein the predefined criterion for sorting objects is a position in the direction perpendicular to the expected orientation of the one of the layer and the boundary. 
     
     
       20. A computer-implemented method of automatically locating a portion of interest in one of an image and matrix data, derived from an item under consideration, comprising:
 identifying parts of the one of the image and the matrix data that satisfy a preset threshold as identified objects possibly included in the portion of interest, the portion of interest corresponding to one of a layer and a boundary, and one of representing and forming part of the item under consideration; 
 applying at least one preselected filter to the one of the image and the matrix data corresponding to the identified objects to find a set of interesting objects most likely to be part of the portion of interest, the preselected filter filtering out the identified objects having a major axis in a direction perpendicular to an expected orientation of the one of the layer and the boundary; 
 sorting the interesting objects in the set into clusters according to a position in the direction perpendicular to the expected orientation of the one of the layer and the boundary; and 
 identifying, in the one of the image and the matrix data based on a known characteristic of the portion of interest, one of the clusters as corresponding to the portion of interest. 
 
     
     
       21. A method as claimed in  claim 20 , wherein, for image data, the preset threshold comprises a pixel value for at least one color, and pixel data of the image which exceed the pixel value are included in the identified objects.